Follow a claim back to its source.

217 countries and economies, 262 city records, and 6,717 retained numeric observations. Individual fields retain source dates, original references, methods, notes and workbook row locations. City boundaries may be metropolitan or service areas. Sent for recycling measures waste collected for recycling, not recovered output.

Units: Source tonnes/year, kg/person/day, percentages and kilometres. Percentages stored as fractions in the workbook are multiplied by 100; original cells are retained.

Preparation: No national-to-city disaggregation or synthetic city locations. Numeric values outside valid bounds are withheld with the original cell preserved. 2022 modeled baseline and future projections are separate from reported quantities. No missing values replaced with zero; no percentages forced to sum to 100.

• Collection does not prove processing or recovered output.
• Mixed source years and treatment stages may not form a closed balance.
• One out-of-range collection value in Valparaíso is withheld; the original cell remains accessible.
• Source table and codebook dates can disagree; both are retained.

Open the original source · Retained records and definitions

CC BY 4.0 · Acquired 2026-09-21

Cook, Ed, Kremena Ionkova, Perinaz Bhada-Tata, Sonakshi Yadav, and Frank van Woerden. 2026. What a Waste 3.0: Global Snapshot of Solid Waste Management Toward Circularity until 2050. World Bank. Country and city datasets.

Domestic extraction is the gross physical mass extracted within a territory and used for economic activity or further transformed. Includes gross metal ore, not only contained metal. Excludes unused mining overburden and extraction not entering economic use. This is DE, not domestic material consumption (DE + imports − exports) or material footprint. Underlying source series contain estimates/models. The publisher identifies 2022–2024 as estimates based on economic proxies; the 2024 technical annex says some projections enter from 2020 and the last year recommended for establishing statistical correlations is 2019.

Units: Native extraction unit t (metric tonnes), native total extraction per-capita t/cap, and native Population counts of people (export unit field blank; metric explicitly Population). No mass conversion. Material per-capita values are derived from source class tonnes / native population; total per-capita uses the native DE/cap field. Missing data stays null.

Preparation: Public selected-export CSV acquired unchanged. SHA-256: 60779224616b80582c6c49115322a23abef308add755dab83acf88eb3728097b. Wide annual columns 1970–2024 melted to country-year rows. Categories mapped Biomass→biomass, Fossil fuels→fossil, Metal ores→metals, Non-metallic minerals→minerals. Publisher code list from the portal maps country names to ISO3+UN M49 codes. Regional aggregates are excluded from countries and rows; World is retained as WORLD. Nine historical boundary codes are retained only in raw archive, not conflated with current polygons. estimated=true identifies guaranteed 2022–2024 late-year proxy estimates. false does NOT mean a direct observation; historical accounts also contain modelled estimates. Cura<e7>ao source display name repaired to Curaçao; its source ISO3 CUW / M49 531 is unchanged. Native totals/ratios export acquired unchanged: Population, DE/cap, DE, 1970–2024. SHA-256: 3cdbd45f9221bbc3dee437e91ed0e94599e389c887817888780daa758ef87761. In enriched v2 only, all-material total is native DE from totals/ratios if present; total per-capita is native DE/cap. Each class per-capita is class tonnes divided by native population. Population is never inferred, and missing material classes stay null even when a reported total exists. Source reconciliation finding: native World DE totals differ slightly from the sum of four exported classes. Largest absolute difference occurs in 1975 (2,681,998.592 tonnes; approximately 0.00776%). Most differences are source rounding. All are retained without forced balancing. 2024 class sum is 106,971,469,898t versus native total 106,971,469,894.03t.

Open the original source

No explicit machine-readable reuse license identified on the database landing page, portal, or downloaded CSV. Public export available; attribution requested by publisher. Do not label CC BY or public domain. · Acquired 2026-09-19

United Nations Environment Programme, International Resource Panel, Global Material Flows Database. 2024 edition. Retrieved 19 September 2026.

Original end-of-year anthropogenic mass, excluding waste; summed six material columns. Biomass uses source-provided years only, without smoothing or interpolation. Historical values are research estimates. Anthropogenic values after 2015 are the publication's extrapolations; biomass values after 2017 are extrapolations. Low/high, where present, are biomass +/- one standard deviation. Gaps between source years remain gaps; connecting chart segments are guides, not annual observations.

Units: Source teratonnes dry weight, normalized by multiplying by 1e12.

Preparation: Six human-made categories summed; waste excluded; no smoothing/interpolation; joined only on available source years; shipped subset ends 2025.

Open the original source

MIT for authors' data/code repository; Nature article has separate publisher rights. · Acquired 2026-09-19

Elhacham, E., Ben-Uri, L., Grozovski, J. et al. Global human-made mass exceeds all living biomass. Nature 588, 442–444 (2020). doi:10.1038/s41586-020-3010-5.

Published Circularity Metric: secondary material input / total primary + secondary material input. Includes recycling and downcycling. Input share, not the share of all waste recycled. Report 2025 v1.0, data 2021. 2026 edition focuses on economic value loss and continues to reference this 6.9% materials metric.

Units: Percent of total material input; flow figure original Gt normalized to metric tonnes.

Preparation: Exact source rounded numbers transcribed; no precision invented; percent not recalculated from rounded tonnes.

Open the original source

CC BY-SA 4.0; the derived transcription is distributed under the same license with attribution. · Acquired 2026-09-19

Circle Economy (2025). The circularity gap report 2025. Amsterdam: Circle Economy. Tables 1–2, pp. 24–25; Figure 3, p. 25.

GWMO 2024 Figure 7, page 21. Original tonnes reported in thousands, multiplied by 1,000. Keep this baseline separate from the World Bank 2022 baseline. Uncontrolled combines dumping and open burning and cannot be separated from this figure. Displayed percentages are source-rounded.

Units: Source Figure 7 thousand tonnes, multiplied by 1,000; rounded percent retained.

Preparation: Transcribed factual values only, no data interpolation or geographic substitution.

Open the original source

Educational/non-profit reproduction with attribution; report prohibits commercial reproduction without permission. No imagery or report text reproduced in data artifact. · Acquired 2026-09-19

United Nations Environment Programme (2024). Global Waste Management Outlook 2024: Beyond an age of waste – Turning rubbish into a resource. Nairobi. Figure 7, p. 21.

Source annual export-direction records from selected reporters. Use netWgt only when isNetWgtEstimated is explicitly false. HS4 records can be source aggregates (isReported=false, isAggregate=true); that does not make the non-estimated net-weight field synthetic. Retain original-classification, report/aggregate, legacy estimation and provenance flags. Mirror imports are not combined with export records. HS6309 is worn articles and is not necessarily waste. Native reported weights can still disagree across partners: a targeted 2024 Chile-to-China HS2603 mirror check found 2,253,844 tonnes reported by Chile versus 9,210,398 tonnes by China, with both non-estimated. Display reporting side explicitly; no reconciliation is inferred. See quality-notes.json. The importer supplement covers HS 2709 and 2710. Selected importer coverage is incomplete; no national total is extrapolated. All original report/aggregate and estimation flags remain available.

Units: Native netWgt in kilograms; canonical metric tonnes = netWgt / 1,000. Commodity mass is not contained-metal mass.

Preparation: Official reporter/partner M49-to-ISO3 lookup. Exclude world/unsupported partners, self-flows, missing/zero/estimated weights and non-total mode/customs/partner2 dimensions. Retain strictly positive exporter-direction records. Capped 500-row queries are split by commodity before inclusion; parent responses are excluded. No interpolation, dollar-to-weight conversion, mirror blending or global extrapolation. Release 16 also retains positive non-estimated oil import net weights. Official Comtrade reporter codes differ from ISO numeric codes; the active official reporter register resolves the request identifiers. In outgoing views, partner imports fill only origin/product/year groups with no retained exports. Incoming oil uses its own importer reports where available, otherwise partner exports.

Open the original source

United Nations Comtrade usage agreement. Attribution and redistribution conditions apply; no Creative Commons or blanket open-data licence is asserted. https://comtradeplus.un.org/LicenseAgreement · Acquired 2026-09-23

United Nations Statistics Division, UN Comtrade Database, annual merchandise trade, retrieved September 2026. Exact public API URLs, response hashes and source row references are retained in the release manifests.

Generalized modern boundaries. Historical country codes are excluded where no compatible geographic concordance exists.

Units: WGS84 longitude and latitude

Preparation: TopoJSON converted to GeoJSON; centroids generated by d3-geo. Country-code joins use publisher M49 mappings.

Open the original source

Public domain (Natural Earth); world-atlas package ISC. · Acquired 2026-09-19

Made with Natural Earth. Geometry packaged by world-atlas.

Domestic extraction records used materials taken within a territory. Domestic material consumption is extraction + physical imports − physical exports. Material footprint (raw material consumption) allocates upstream extraction to final demand using a multiregional input-output model. These are distinct accounts. Preserve the published native values: world direct imports/exports do not balance and native MF does not exactly equal the DE/RME balance. Do not turn that difference into an invented flow. Years 2022–2024 are source proxy estimates; earlier years include modelling too.

Units: Native t (metric tonnes), t/cap (tonnes/person), and Population (people). No mass conversion. Four-class footprint per capita = reported class tonnes / native IRP population; total per capita is native MF/cap.

Preparation: Melt annual CSV columns. Preserve null gaps. Join portal country names using publisher ISO3+M49 codes and existing atlas country contract; exclude nine obsolete countries and all regional aggregates except WORLD. Keep negative DMC/PTB as reported. Never sum modern countries as a world total. MF class values have integer tonne precision; totals/ratios retain publisher precision. No interpolation, stock estimation or bilateral inference. Source class sums are not forced to match independently reported totals.

• Raw MF accounting differences are retained, not balanced.
• 47 negative native DMC observations remain signed.
• Global MF has fewer explicitly separate countries than DE; Rest-of regions are excluded from country selections.
• Latest years are estimates, not live measurement.
• The four-class MF export can differ from native total MF; the largest absolute difference is 394,825,439.862 tonnes (WORLD 2002). Preserve both; do not force balancing or describe every difference as rounding.

Open the original source · Retained records and definitions

Public export. No explicit reuse licence identified on the reviewed IRP pages/CSV; attribution required. No Creative Commons licence inferred. · Acquired 2026-09-19

UNEP International Resource Panel (2024). Global Material Flows Database. Compiled by CSIRO and WU Vienna. Retrieved 19 September 2026.

Source-reported production of eight selected primary crops: wheat, rice, barley, maize, potatoes, sugar cane, soya beans and oil palm fruit. Editorial subset, not total agricultural output or biomass extraction. Retains official/estimated/imputed flags.

Units: Source production quantities in t, retained as metric tonnes. Crop-specific commodity moisture basis remains the FAO basis; no biomass dry-mass conversion.

Preparation: Select QCL Production rows with source unit t and eight explicit item codes. Join source M49 to existing atlas countries; preserve WORLD and exclude historical/regional aggregates. Retain exact numeric values and flags. Archive 1970–2024 subset; publish only exact 2024 rows, without fallback to another year.

• Selected crops only; never label as biomass total.
• Primary crop tonnes differ in moisture basis and are not additive with processed agricultural output.
• No nearest-year substitution; missing national values stay absent.

Open the original source · Retained records and definitions

CC BY 4.0 plus FAO Statistical Database Terms of Use; attribution, no endorsement, and applicable third-party exceptions. · Acquired 2026-09-19

FAO. 2025. FAOSTAT: Crops and livestock products. Accessed 19 September 2026. https://www.fao.org/faostat/en/#data/QCL. Licence: CC BY 4.0.

1,388 production records for nine forest products. Roundwood, sawnwood, panels, pellets, wood pulp, other-fibre pulp, recovered paper and paper/board retain product-specific units and official/estimated/imputed source flags. FAO may estimate missing submissions or repeat past figures until new information is available.

Units: Cubic metres and tonnes by product. Do not convert timber volume to tonnes without a documented density basis.

Preparation: Select exact 2024 Production elements for nine explicit item codes. Retain m3 as cubic metres and t as metric tonnes; no density conversion or aggregation across processing stages. Map source M49 only to known countries; retain the source World aggregate.

• Nine selected products, not a sum of all forest output.
• Raw and processed products overlap and must not be added together.
• Volume and mass remain separate units.

Open the original source

CC BY 4.0 plus FAO Statistical Database Terms of Use; verify third-party exceptions and retain attribution/no endorsement. · Acquired 2026-09-21

FAO. 2026. FAOSTAT, Forestry Production and Trade (FO). Bulk release updated 9 January 2026. Acquired 21 September 2026. Selected 2024 Production records. Licence: CC BY 4.0; FAO Statistical Database Terms of Use apply.

349,724 source observations. All native observations from 2023 onward. Scanned 4,820,497 original CSV rows. Former territories and regional aggregates retain FAO identities and are not mapped as countries. No interpolation, conversion or reconciliation.

Units: Commodity-specific tonnes, nutrient/energy and per-person measures; preserve the element unit.

Preparation: Native values and dimensions preserved. Missing observations are not replaced with zero.

Open the original source · Retained records and definitions

CC BY 4.0 plus FAO Statistical Database Terms of Use; verify third-party exceptions and retain attribution/no endorsement. · Acquired 2026-09-30

FAO. Food Balances (FBS). Bulk metadata updated 28 October 2025. Source inspected 19 September 2026.

859,928 source observations. All native observations from 2023 onward. Scanned 11,615,212 original CSV rows. Former territories and regional aggregates retain FAO identities and are not mapped as countries. No interpolation, conversion or reconciliation.

Units: Source item/element units including tonnes; processed outputs must not be added to raw input totals.

Preparation: Native values and dimensions preserved. Missing observations are not replaced with zero.

Open the original source · Retained records and definitions

CC BY 4.0 plus FAO Statistical Database Terms of Use; verify third-party exceptions and retain attribution/no endorsement. · Acquired 2026-09-30

FAO. Supply Utilization Accounts (SCL). Bulk metadata updated 1 November 2025. Source inspected 19 September 2026.

2,060,725 source observations. All native observations from 2023 onward; quantity rows only, reporter and partner perspectives separate. Scanned 52,410,630 original CSV rows. Former territories and regional aggregates retain FAO identities and are not mapped as countries. No interpolation, conversion or reconciliation.

Units: Native units; no conversion applied because no numerical observations are bundled.

Preparation: Native values and dimensions preserved. Missing observations are not replaced with zero.

Open the original source · Retained records and definitions

FAOSTAT data licensing and source-specific conditions require verification with acquired release. · Acquired 2026-09-30

FAO. FAOSTAT, Detailed trade matrix (TM). Catalogue reviewed 19 September 2026.

40,879 source observations. All 2023–2024 rows from the March 2026 global production workspace. Capture and aquaculture remain separate; FAO fishing area is a reporting area, not a vessel location. Q_tlw is tonnes of live weight (aquatic plants: wet weight); numerical organism counts retain their native measure code. No conversion into landed product weight or trade shipments.

Units: Source series include live-weight tonnes and product-specific weights.

Preparation: Native values and dimensions preserved. Missing observations are not replaced with zero.

Open the original source · Retained records and definitions

CC BY 4.0 plus FAO Statistical Database Terms of Use; verify third-party exceptions and retain attribution/no endorsement. · Acquired 2026-09-30

FAO. Fisheries and Aquaculture statistics. Current corporate statistical database. Source inspected 19 September 2026.

Acquisition note: Full machine-readable release and metadata remain inaccessible: official ScienceBase and USGS data endpoints returned HTTP 403 on 2026-09-30. The full 90+ commodity release is not imported.

116 country observations across copper mine/refinery, aluminium smelter, lithium mine, raw steel, silicon metal, ferrosilicon and salt. 2025 estimate columns transcribed from primary USGS printed tables; source PDF hashes retained.

Units: Commodity-specific units (including metric tonnes and contained-metal tonnes); no universal ore conversion.

Preparation: Printed thousands converted to metric tonnes. Withheld U.S. silicon excluded; printed silicon dashes are genuine zeroes.

Open the original source · Retained records and definitions

USGS-authored federal data are public domain unless marked otherwise; third-party content retains its rights. · Acquired 2026-09-30

U.S. Geological Survey. Mineral Commodity Summaries 2026; Minerals Yearbook. 2026 edition. Source inspected 19 September 2026.

Acquisition note: BGS terms require permission for commercial use or supplying data to third parties. No redistribution permission is available for this public atlas.

Country commodity production for over 70 mineral commodities. API suppresses some publication symbols into null/zero; do not interpret every null as missing or every zero as measured nil.

Units: Commodity-specific source units and ore/metal bases.

Preparation: No quantitative transformation or browser data published from this catalog entry.

Open the original source

BGS terms permit specified non-commercial research uses; commercial use or supply to third parties requires checking permission with the rights owner. · Acquired 2026-09-19

British Geological Survey. World Mineral Statistics. World Mineral Production 2020–2024. Source inspected 19 September 2026.

29,862 source observations. Native JSON-stat dimensions, units and status flags. Regional aggregates are separately prefixed ESTAT and excluded from country maps. No unit conversion or addition to IRP accounts. Kosovo cells withheld under publisher exception.

Units: Selected unit must be validated; thousand tonnes must be multiplied by 1,000 for normalized tonnes.

Preparation: Native values and dimensions preserved. Missing observations are not replaced with zero.

Open the original source · Retained records and definitions

Eurostat reuse with attribution, including commercial reuse, subject to its geographic and third-party exceptions: https://ec.europa.eu/eurostat/web/main/help/copyright-notice . Public outputs exclude non-EU/EFTA/official-candidate country records covered by those exceptions. · Acquired 2026-09-30

European Commission / Eurostat. env_ac_mfa; env_ac_mfadpo; env_ac_mfabi. Current Eurostat datasets. Source inspected 19 September 2026.

546 source observations. Native JSON-stat dimensions, units and status flags. Regional aggregates are separately prefixed ESTAT and excluded from country maps. No unit conversion or addition to IRP accounts. Kosovo cells withheld under publisher exception.

Units: Source unit selection required; retain model and scale metadata.

Preparation: Native values and dimensions preserved. Missing observations are not replaced with zero.

Open the original source · Retained records and definitions

Eurostat reuse with attribution, including commercial reuse, subject to its geographic and third-party exceptions: https://ec.europa.eu/eurostat/web/main/help/copyright-notice . Public outputs exclude non-EU/EFTA/official-candidate country records covered by those exceptions. · Acquired 2026-09-30

European Commission / Eurostat. Material footprints / raw material equivalents (env_ac_rme). Current Eurostat dataset. Source inspected 19 September 2026.

121,304 source observations. Versioned dataflow 1.0, all returned 2020–2024 observations. Unit multiplier is preserved separately: T with multiplier 6 means millions of tonnes. No silent scale conversion. Per-capita ratios and monetary productivity are separate units; missing values stay null. OECD aggregates retain prefixed identities. Accounts overlap IRP and Eurostat and must not be added.

Units: Indicator page uses tonnes/person; underlying table contains metric-specific units.

Preparation: Native values and dimensions preserved. Missing observations are not replaced with zero.

Open the original source · Retained records and definitions

OECD data terms permit reuse with credit, including commercial use, except additional dataset/third-party restrictions. · Acquired 2026-09-30

OECD. Material resources / material consumption. Current OECD Data Explorer. Source inspected 19 September 2026.

4,140 source observations. Complete returned regional and world series under indicator 12.2.1, 2000–2022. Regional estimates retain UN identities; no synthetic country allocation. Goal 8 duplicates excluded. Per-capita and GDP ratios retain their denominator and are not mass totals.

Units: Metric and per-capita/per-GDP series have different units; select exact series code and units.

Preparation: Native values and dimensions preserved. Missing observations are not replaced with zero.

Open the original source · Retained records and definitions

UN/UNEP source-specific terms; no Creative Commons licence inferred. · Acquired 2026-09-30

UN Statistics Division / UNEP. SDG 12.2.1 and 12.2.2 (also 8.4.1/8.4.2). Metadata updated 28 March 2025. Source inspected 19 September 2026.

Acquisition note: Catalogued. Current license does not establish permission for Gaia commercial-product redistribution.

Industry/product environmental input-output model. Monetary transaction matrices are not shipment-tonne data. Material extensions overlap IRP inputs and require matrix calculations to form consumption footprints.

Units: Economic core: million EUR; environmental extensions have independent units.

Preparation: No quantitative transformation or browser data published from this catalog entry.

Open the original source

Customized non-commercial academic CC-BY-SA-NC derivative for v3.9+; commercial license separate. Latest record supersedes stale general-homepage license wording. · Acquired 2026-09-19

EXIOBASE consortium / XIO Sustainability Analytics. EXIOBASE 3 EE-MRIO. 3.10.2; published 13 May 2026. Source inspected 19 September 2026.

Acquisition note: No licensed credential available. Do not acquire or redistribute restricted tables.

MRIO model with national industry classifications and a reduced environmental satellite set compared with Eora1. Monetary flows cannot be represented as physical shipments.

Units: Economic-account and satellite-specific units; require licensed metadata.

Preparation: No quantitative transformation or browser data published from this catalog entry.

Open the original source

Active subscription/license required; eligible academic use has separate free license. Eora1 licenses do not transfer. · Acquired 2026-09-19

KGM & Associates / Eora. Eora2 Global Supply Chain Database. Current Eora2 release. Source inspected 19 September 2026.

36,576 source observations. Complete four numerical worksheets of the official R2024A V1-0 country workbook. Producer-aggregated land area, population, built-up surface and built-up surface per capita by three settlement classes, 1975–2030 in five-year intervals. 2025/2030 are projections; prior years are modeled estimates. Native numerical values are preserved, including fractional population estimates and genuine zeroes. Publisher dashes stay null with their marker. Uses producer GADM 4.1 geography; no boundary geometry is republished. Context quantities do not measure material mass, annual extraction or verified facility throughput. Attribution: Carioli et al., European Commission Joint Research Centre, GHS-COUNTRY-STATS R2024A. Transformation: workbook cells converted to partitioned JSON without numerical conversion.

Units: People per cell; square metres; cubic metres. No conversion to tonnes without a separate calibrated model.

Preparation: Native values and dimensions preserved. Missing observations are not replaced with zero.

Open the original source · Retained records and definitions

Open reuse with source acknowledgement under the published GHSL terms. · Acquired 2026-09-30

European Commission Joint Research Centre. GHS-POP, GHS-BUILT-S/V; GHS-WUP population and built-up projections. R2023A and R2025A. Source inspected 19 September 2026.

Acquisition note: Global2 R2025A v1 population-estimates ZIP is 5,584,195,277 bytes. Two bulk transfers stopped before completion; explicit byte-range resume returned the full file (HTTP 200). No partial archive is counted as a complete import.

Gridded demographic estimates for local exposure and per-person analysis. Select a consistent population product/year; do not substitute national gaps using regional values.

Units: Population counts or density as defined by each raster.

Preparation: No quantitative transformation or browser data published from this catalog entry.

Open the original source

CC BY 4.0, with dataset-specific citation. · Acquired 2026-09-19

WorldPop / University of Southampton. WorldPop population counts and density. Current WorldPop Hub. Source inspected 19 September 2026.

Acquisition note: Catalogued. Resolve redistribution terms and select physical-mass series before integration.

Production, trade, conversion and end use of fuels, electricity and heat. Fossil fuel mass is only a subset; energy quantities do not equal tonnes.

Units: Native metric tonnes, volume, GWh or terajoules depending on series. Preserve original units.

Preparation: No quantitative transformation or browser data published from this catalog entry.

Open the original source · Retained records and definitions

UNSD permits attributed non-profit reuse; other uses require contacting UNSD. Full archive is separately distributed. · Acquired 2026-09-19

United Nations Statistics Division. Energy Statistics Database / UNdata DF_UNDATA_ENERGY. Current UNSD annual release. Source inspected 19 September 2026.

Acquisition note: Landing page reachable but metadata did not render in the research tool; no specific archive or values acquired.

Candidate high-resolution MRIO for upstream resource attribution. Keep independent from direct physical shipment statistics and from the existing IRP national accounts.

Units: Economic matrix and satellite units must be acquired from release metadata.

Preparation: No quantitative transformation or browser data published from this catalog entry.

Open the original source

License not verified for a specific release. · Acquired 2026-09-19

IELab / University of Sydney. Global Resource Input-Output Assessment (GLORIA). Release not pinned. Source inspected 19 September 2026.

Company-reported tailings disclosures.2,144 rows acquired;193 markedduplicates and31 missing/invalid coordinates leave1,920 distinct geolocated records. Homepage1,805 count differs from actual current API.

Units: Storage volume m³, height m. Never convert volume to tonnes without density.

Preparation: Schema/adapter and raw snapshot present; excluded from public site artifacts pending license resolution. No invented dates, no risk ranking from incomparable hazard classifications.

Open the original source

No explicit open bulk reuse license found in portal terms https://tailing.grida.no/gdpr . Public viewing is not treated as a redistribution grant. · Acquired 2026-09-19

GRID-Arendal, Global Tailings Portal. Public map APIs taillingLoc and tailings_all, retrieved2026-09-19.

Acquisition note: The public explorer lists 7,108 mines across all statuses, but CSV export requires full name, email, organization, sector, intended use and explicit licence acknowledgement. No form submitted.

Tracked operating, proposed and retired mines; usual threshold1Mt/year, China0.45Mt/year. Source may substitute capacity when production is missing. Must preserve production versus capacity and location accuracy.

Units: Million tonnes production; Mt/year capacity.

Preparation: Catalogued and methodology verified; complete source file not acquired. Download flow requests user details. No form submitted and no values incorporated.

Open the original source

GEM Creative Commons Public License; exact downloadable artifact terms require inspection before publication. · Acquired 2026-09-19

Global Energy Monitor, Global Coal Mine Tracker, August2026 release.

Primary NGO documentation identifies Alang-Sosiya, Chattogram/Sitakund,Gadani and Turkish sites. No georeferenced bulk yard inventory acquired.

Units: Ship counts and various ship tonnage concepts; gross tonnage is not mass.

Preparation: Destination geography documented but no invented exact coordinates. OSM region queries timed out; no guessed shipyard markers rendered.

Open the original source

No blanket open database license confirmed. Country-profile text not republished; underlying factual destinations catalogued. · Acquired 2026-09-19

NGO Shipbreaking Platform, country profiles India,Bangladesh,Pakistan and annual shipbreaking lists.

Acquisition note: The broad Overpass query returned HTTP 406; a focused western-Canada query timed out. No complete source export is acquired.

Volunteer mapping. landuse=landfill is geography, not proof of operational status, waste mass, safety or complete coverage. General shipyard tags include ship construction/repair and must not be relabeled shipbreaking.

Units: Coordinates and geometry; no universal mass values.

Preparation: Adapter approach catalogued. Initial exact industrial=shipbreaking tag query returned zero; broader name queries timed out. An unrelated candidate node was rejected by coordinate validation. No OSM points added.

Open the original source

ODbL1.0 for database; attribution and share-alike obligations apply. · Acquired 2026-09-19

© OpenStreetMap contributors. Data available under the Open Database License.

2,678,045 observations across all 16 native EEA v16 CSV tables. Complete native rows from F6_1_IED_Installations.csv in the public February 2026 user-friendly CSV package. All original fields are preserved as native_ fields, including coordinates, identifiers, status, confidentiality reasons and sentinel date strings. National, sector/activity and facility perspectives are separate tables and overlap; do not add them. Missing/confidential numeric cells stay null. Waste transfers are annual off-site tonnes, pollutant releases/transfers kg/year; combustion measures retain their own reported units. Administrative and multi-field incineration capacity rows have null generic value; capacity is in the native fields and is not throughput. Coordinates do not establish a shipment route or identify a receiving waste facility. Country reporting-year histories can contain repeated site identities and are not counts of unique active sites. Attribution: European Environment Agency, Industrial Reporting v16, DOI 10.2909/657ac3cb-affa-4295-a4a9-27b4f539adab. Changes: original CSV rows converted to JSON and partitioned into at most 5,000-row segments; no numerical scale conversion.

Units: Pollutant tonnes, waste-transfer tonnes and facility coordinates; taxonomies remain native.

Preparation: Native values and dimensions preserved. Missing observations are not replaced with zero.

Open the original source · Retained records and definitions

Release-specific EEA metadata terms need verification for each acquired dataset. · Acquired 2026-09-30

European Environment Agency, European Industrial Emissions Portal.

1,626,092 source observations. Complete facility member of the acquired September 2026 RCRAInfo ZIP; partitioned by native state code. 2026 is the acquisition edition, not facility operation year. All 15 original fields retain exact strings, including status and generator codes, coordinate strings and blank coordinates. A registration does not establish current operating status, waste quantity or shipment connections. The value column is null because throughput is not reported in this facility table. Enforcement and violation members are outside this retained location-register scope.

Units: Facility coordinates; waste types and metrics dataset-specific.

Preparation: Native values and dimensions preserved. Missing observations are not replaced with zero.

Open the original source · Retained records and definitions

EPA geospatial public-domain default; each dataset metadata must be retained. · Acquired 2026-09-30

US EPA, Enforcement and Compliance History Online Data Downloads.

Mineral occurrences, prospects and mine records require status/category distinction.

Units: Locations and native commodity/resource metadata.

Preparation: Catalogued; primary endpoint returnedHTTP403 in this environment. No data added.

Open the original source

US government-produced data generally public domain; release metadata not retrieved. · Acquired 2026-09-19

US Geological Survey, Mineral Resources Data System.

UNITAR/WESR model estimates and national/Eurostat observations as identified on each row. UN SDG series is not the complete GEM country table. UNITAR values after2022 are labelled model projections; missing countries remain unavailable.

Units: Metric tonnes for mass; percentages and tonnes/person where specified. Original values and units retained.

Preparation: Parse complete Series/Data JSON response; map UN M49 to ISO3; kilograms/person÷1000; tonnes and percentages unchanged. Exclude nonfinite missing observations, preserve footnotes/flags/dimensions. Keep regions in a separate artifact.

Open the original source · Retained records and definitions

UN statistical-data terms; no explicit Creative Commons licence supplied by this API. Original agency and country attribution retained per observation. GEM report and photographs have separate, more restrictive terms. · Acquired 2026-09-19

UN Statistics Division / UNEP / UNITAR / national statistical offices (2026), Electronic waste generated and recycled, 2026.Q2.G.02 API snapshot, 19 September 2026. Retrieved 19 September 2026.

National questionnaires and Eurostat/OECD figures compiled by UNSD/UNEP. National definitions and reporting years differ. Keep country footnotes and observation status. Never add these quantities to World Bank figures for the same waste stream.

Units: Metric tonnes for mass; percentages and tonnes/person where specified. Original values and units retained.

Preparation: Parse complete Series/Data JSON response; map UN M49 to ISO3; kilograms/person÷1000; tonnes and percentages unchanged. Exclude nonfinite missing observations, preserve footnotes/flags/dimensions. Keep regions in a separate artifact.

Open the original source · Retained records and definitions

UN statistical-data terms; no explicit Creative Commons licence supplied by this API. Original agency and country attribution retained per observation. GEM report and photographs have separate, more restrictive terms. · Acquired 2026-09-19

UN Statistics Division / UNEP / UNITAR / national statistical offices (2026), Municipal waste generation and recycling, 2026.Q2.G.02 API snapshot, 19 September 2026. Retrieved 19 September 2026.

National and Eurostat/UNSD observations. Treatment categories retain source taxonomy: incineration, incineration with energy recovery, landfill, recycling and other management. Imported/exported treatment can differ from domestic generation; no forced mass closure.

Units: Metric tonnes for mass; percentages and tonnes/person where specified. Original values and units retained.

Preparation: Parse complete Series/Data JSON response; map UN M49 to ISO3; kilograms/person÷1000; tonnes and percentages unchanged. Exclude nonfinite missing observations, preserve footnotes/flags/dimensions. Keep regions in a separate artifact.

Open the original source · Retained records and definitions

UN statistical-data terms; no explicit Creative Commons licence supplied by this API. Original agency and country attribution retained per observation. GEM report and photographs have separate, more restrictive terms. · Acquired 2026-09-19

UN Statistics Division / UNEP / UNITAR / national statistical offices (2026), Hazardous waste generation and treatment, 2026.Q2.G.02 API snapshot, 19 September 2026. Retrieved 19 September 2026.

UNEP Food Waste Index2021/2024 estimates, alongside national observations. Household, retail, out-of-home and total sectors are distinct metrics. Confidence footnotes remain visible; national totals may be model estimates, not direct measurements. No addition across total and components.

Units: Metric tonnes for mass; percentages and tonnes/person where specified. Original values and units retained.

Preparation: Parse complete Series/Data JSON response; map UN M49 to ISO3; kilograms/person÷1000; tonnes and percentages unchanged. Exclude nonfinite missing observations, preserve footnotes/flags/dimensions. Keep regions in a separate artifact.

Open the original source · Retained records and definitions

UN statistical-data terms; no explicit Creative Commons licence supplied by this API. Original agency and country attribution retained per observation. GEM report and photographs have separate, more restrictive terms. · Acquired 2026-09-19

UN Statistics Division / UNEP / UNITAR / national statistical offices (2026), Food Waste Index country and sector data, 2026.Q2.G.02 API snapshot, 19 September 2026. Retrieved 19 September 2026.

FAO loss before retail, by food-product group. Global and regional estimates; no national numeric data in this acquired series. Food loss and retail/household food waste have different boundaries and denominators; never sum their percentages.

Units: Metric tonnes for mass; percentages and tonnes/person where specified. Original values and units retained.

Preparation: Parse complete Series/Data JSON response; map UN M49 to ISO3; kilograms/person÷1000; tonnes and percentages unchanged. Exclude nonfinite missing observations, preserve footnotes/flags/dimensions. Keep regions in a separate artifact.

Open the original source · Retained records and definitions

UN statistical-data terms; no explicit Creative Commons licence supplied by this API. Original agency and country attribution retained per observation. GEM report and photographs have separate, more restrictive terms. · Acquired 2026-09-19

UN Statistics Division / FAO (2026), Food loss percentages, 2026.Q2.G.02 API snapshot, 19 September 2026. Retrieved 19 September 2026.

Share of circular material use in overall material use. Eurostat indicator denominator differs from the global Circularity Gap Report; do not rank both as if methodologically identical.

Units: Metric tonnes for mass; percentages and tonnes/person where specified. Original values and units retained.

Preparation: Decode JSON-stat dimensions; preserve source flags. Thousand tonnes×1000 where applicable, tonnes and rates unchanged. EU/other aggregates stored separately from national data.

Open the original source · Retained records and definitions

Eurostat reuse with attribution, including commercial reuse, subject to its geographic and third-party exceptions: https://ec.europa.eu/eurostat/web/main/help/copyright-notice . Public outputs exclude non-EU/EFTA/official-candidate country records covered by those exceptions. · Acquired 2026-09-19

Eurostat (2025), Circular material use rate, JSON-stat API snapshot; updated2025-11-24T23:00:00+0100. Retrieved 19 September 2026.

National municipal waste statistics. GEN and TRT totals overlap their child categories. DSP_I_RCV_E overlaps DSP_I and RCV_E; RCY overlaps RCY_M and RCY_C_D. Render source-selected categories without summing parents and children.

Units: Metric tonnes for mass; percentages and tonnes/person where specified. Original values and units retained.

Preparation: Decode JSON-stat dimensions; preserve source flags. Thousand tonnes×1000 where applicable, tonnes and rates unchanged. EU/other aggregates stored separately from national data.

Open the original source · Retained records and definitions

Eurostat reuse with attribution, including commercial reuse, subject to its geographic and third-party exceptions: https://ec.europa.eu/eurostat/web/main/help/copyright-notice . Public outputs exclude non-EU/EFTA/official-candidate country records covered by those exceptions. · Acquired 2026-09-19

Eurostat (2026), Municipal waste by management operation, JSON-stat API snapshot; updated2026-03-30T23:00:00+0200. Retrieved 19 September 2026.

Packaging material and operation taxonomy preserved, including plastics, paper/cardboard, metals, wood and glass. W1501 total includes components. Recycling national/EU/nonEU destination categories overlap total recycling.

Units: Metric tonnes for mass; percentages and tonnes/person where specified. Original values and units retained.

Preparation: Decode JSON-stat dimensions; preserve source flags. Thousand tonnes×1000 where applicable, tonnes and rates unchanged. EU/other aggregates stored separately from national data.

Open the original source · Retained records and definitions

Eurostat reuse with attribution, including commercial reuse, subject to its geographic and third-party exceptions: https://ec.europa.eu/eurostat/web/main/help/copyright-notice . Public outputs exclude non-EU/EFTA/official-candidate country records covered by those exceptions. · Acquired 2026-09-19

Eurostat (2026), Packaging waste by management operation, JSON-stat API snapshot; updated2026-05-11T23:00:00+0200. Retrieved 19 September 2026.

Source unit RT_TGT2025: rates under2025-and-onwards target methodology, by packaging type. Historical break flags and source precision preserved.

Units: Metric tonnes for mass; percentages and tonnes/person where specified. Original values and units retained.

Preparation: Decode JSON-stat dimensions; preserve source flags. Thousand tonnes×1000 where applicable, tonnes and rates unchanged. EU/other aggregates stored separately from national data.

Open the original source · Retained records and definitions

Eurostat reuse with attribution, including commercial reuse, subject to its geographic and third-party exceptions: https://ec.europa.eu/eurostat/web/main/help/copyright-notice . Public outputs exclude non-EU/EFTA/official-candidate country records covered by those exceptions. · Acquired 2026-09-19

Eurostat (2026), Packaging recycling rates, JSON-stat API snapshot; updated2026-05-11T23:00:00+0200. Retrieved 19 September 2026.

Economy-wide dynamic, inflow-driven stock-flow model. Four end uses × four material groups. Values are model estimates; uncertainty is not supplied in this summarized CSV. Stock is accumulated mass, not annual extraction.

Units: Metric tonnes for mass; percentages and tonnes/person where specified. Original values and units retained.

Preparation: Kilotonnes×1000 to metric tonnes. Sum exactly16 disjoint end-use/material cells per country/process/year using compensated summation. Latest2016 totals in metrics; complete1900–2016 totals in stock-history; components retained separately.

Open the original source

CC BY4.0 (dataset). Model software has a separate GPL licence. · Acquired 2026-09-19

Wiedenhofer et al. / BOKU University (2024), MAT_STOCKS: national material stocks and flows, v1.0, Zenodo12794253. Retrieved 19 September 2026.

Detailed stock, gross additions and end-of-life flows across12end uses and19materials. This is1990–2016, not the later EEA2022extension.

Units: Metric tonnes for mass; percentages and tonnes/person where specified. Original values and units retained.

Preparation: Raw CSV acquired. Retained for material-specific exploration; summarized global artifact already includes these27countries, so no duplicate import.

Open the original source

CC BY4.0 · Acquired 2026-09-19

Wiedenhofer et al. / BOKU University (2025), MAT_STOCKS detailed EU27 extract, v1.0, Zenodo15090142. Retrieved 19 September 2026.

ENV-Linkages model and plastics lifecycle estimates. Rounded values transcribed from the official executive summary/press release. Model mass/fate dataset is published separately under oecd-plastics-database-2022.

Units: Metric tonnes for mass; percentages and tonnes/person where specified. Original values and units retained.

Preparation: Million tonnes×1e6; published percentage shares unchanged. No deriving tonnes from rounded percentages.

Open the original source

OECD Terms and Conditions: data may be reused including commercially with attribution, subject to third-party restrictions. · Acquired 2026-09-19

OECD (2022), Global Plastics Outlook: published global estimates, Revised April2022. Retrieved 19 September 2026.

Historical model estimates,1990–2019. Five separate waste fates exactly sum to source total. Reported geographic model regions do not generally match countries: China includes Hong Kong. Only Canada, USA and India are mapped to national ISO3; WORLD is the native source global total. Regional model entities stay separate.

Units: Metric tonnes for mass; percentages and tonnes/person where specified. Original values and units retained.

Preparation: Parse publicly embedded backend JSON without executing code. Million tonnes×1e6. Preserve five source fate categories and exact native values. No post2019 scenario projections imported into factual metrics.

Open the original source

OECD Terms and Conditions, section3: attribution; permitted reuse including commercial, subject to third-party restrictions. · Acquired 2026-09-19

OECD (2022), Global Plastics Outlook: use, fate, leakage and stocks, OECD Compare Your Country embedded data snapshot. Retrieved 19 September 2026.

Country/territory2022 generation and documented formal collection/recycling. The report mixes estimated generation and documented collection; missing collection is N/A, not zero.

Units: Metric tonnes for mass; percentages and tonnes/person where specified. Original values and units retained.

Preparation: PDF acquired for methodology/verification. Bulk country table not imported into browser due to its distinct reuse conditions; public SDG country observations used instead.

Open the original source

CC BY-NC-SA3.0IGO, excluding third-party photographs/material. Commercial reuse may need permission. · Acquired 2026-09-19

UNITAR / ITU (2024), Global E-waste Monitor2024, Edition2, November2024. Retrieved 19 September 2026.

39,503 source observations. Complete numerical study observations returned by the anonymous FAO export with 1965–2026 selected, all countries, commodities, stages and methods, including broadly defined product groups. Actual returned years 1966–2026. Each row is a literature/study observation and can overlap other rows. Values are native loss percentages, not national annual mass, food-waste tonnage or an official FAO endorsement. Original percentage strings, sampling methods, source references, URLs and study-region fields remain available. Loss quantities remain exact strings: the export mixes missing, numeric and text quantities and does not provide a uniform mass unit. No averaging, mass conversion, country scaling or imputation. Narrative notes omitted; the source export is separately reachable through the publisher.

Units: Metric tonnes for mass; percentages and tonnes/person where specified. Original values and units retained.

Preparation: Native values and dimensions preserved. Missing observations are not replaced with zero.

Open the original source · Retained records and definitions

Underlying literature/databases retain their own rights; FAO warns against assuming endorsement or accuracy. · Acquired 2026-09-30

FAO (2026), Food Loss and Waste Database, Live2026portal. Retrieved 19 September 2026.

147,776 source observations. Native JSON-stat dimensions, units and status flags. Regional aggregates are separately prefixed ESTAT and excluded from country maps. No unit conversion or addition to IRP accounts. Kosovo cells withheld under publisher exception.

Units: Metric tonnes for mass; percentages and tonnes/person where specified. Original values and units retained.

Preparation: Native values and dimensions preserved. Missing observations are not replaced with zero.

Open the original source · Retained records and definitions

Eurostat reuse with attribution, including commercial reuse, subject to its geographic and third-party exceptions: https://ec.europa.eu/eurostat/web/main/help/copyright-notice . Public outputs exclude non-EU/EFTA/official-candidate country records covered by those exceptions. · Acquired 2026-09-30

Eurostat (n.d.), Waste generation by economic activity, Live database. Retrieved 19 September 2026.

Final treatment inside the reporting country, including imported waste and excluding exported waste. It is not the fate of that country's generated waste. Pre-treatment such as sorting and drying is excluded. Sludges and dredging spoils are measured as dry matter. TOTAL and TOT_X_MIN are separate, overlapping views. TOT_X_MIN excludes major mineral waste; never add the two scopes together. TRT is a total. RCV_R_B combines recycling and backfilling; DSP_L_OTH combines landfill and other disposal. Never add totals/combined categories to their children. Prefer the six separate operations. Where split cells are absent, use the source's combined category, or show the gap; do not derive a confidential split by subtraction. Country observations can contain rounded quantities; EU aggregates are rounded to 10,000 tonnes. Component sums may differ from the native total; totals are not adjusted. This is all-sector waste treatment, not municipal solid waste alone. The latest available reference year is 2022; no 2024 values were available in this snapshot.

Units: Native unit T (metric tonne), factor 1. Original values, units, sparse missing cells and status flags retained. No conversion from money, capacity or counts.

Preparation: Filter native T mass unit; map Eurostat country codes to ISO3; retain source cell IDs and raw JSON-stat indices; preserve EU aggregates under their own codes; exclude restricted geographies. Native hazardous + non-hazardous total only; separate TOTAL and TOT_X_MIN views; no subgroup sum or invented residual.

Open the original source

Eurostat reuse with attribution, including commercial reuse, subject to its geographic and third-party exceptions: https://ec.europa.eu/eurostat/web/main/help/copyright-notice . Public outputs exclude non-EU/EFTA/official-candidate country records covered by those exceptions. · Acquired 2026-09-19

Eurostat, env_wastrt, DOI: 10.2908/ENV_WASTRT. Accessed 2026-09-19. Adapted by OVERSHOOT.

Open-scope WEEE reporting, six equipment categories. Only reference years 2018 onwards are included here; reporting before 2019 has a transitional scope. EE6 is the source total. Large equipment includes photovoltaic panels; never sum EE_LE with its EE_LEXPVP/EE_LE_PVP children. TRT_NAT, TRT_EU_FOR and TRT_NEU describe treatment in the reporting Member State, another EU Member State, and outside the EU. Specific receiving countries/facilities are not identified. These treatment locations are reported for WEEE associated with the reporting country's collection system. They are not inferred from UN Comtrade shipments. Recovery, recycling and preparing for reuse are overlapping performance categories. Never add RCV, RCY_PRP_REU, RCY and PRP_REU as mutually exclusive fates. Missing destination cells remain absent, including countries that report only treatment at home. Missing is not zero, and no residual is labelled informal disposal. Separate collection and treatment figures do not describe every item of e-waste generated; uncollected material is not assigned a destination. Country-specific methodological notes apply: Germany's national-treatment cell includes confidential exports, while Hungary reports first treatment. These are not verified ultimate disposal destinations.

Units: Native unit T (metric tonne), factor 1. Original values, units, sparse missing cells and status flags retained. No conversion from money, capacity or counts.

Preparation: Filter native T mass unit; map Eurostat country codes to ISO3; retain source cell IDs and raw JSON-stat indices; preserve EU aggregates under their own codes; exclude restricted geographies. Restrict to 2018 onwards; publish EE6 total in the browser and retain detailed categories in Parquet. Collection, recovery and treatment totals are not summed.

Open the original source

Eurostat reuse with attribution, including commercial reuse, subject to its geographic and third-party exceptions: https://ec.europa.eu/eurostat/web/main/help/copyright-notice . Public outputs exclude non-EU/EFTA/official-candidate country records covered by those exceptions. · Acquired 2026-09-19

Eurostat, env_waseleeos, DOI: 10.2908/ENV_WASELEEOS. Accessed 2026-09-19. Adapted by OVERSHOOT.

Acquisition note: Complete five-value Figure 4.6 workbook retained. Remaining country and scenario tables are not acquired; the publisher website kept this browser in a security-verification loop on 2026-09-30.

5 source observations. Complete five numerical cells from Figure 4.6, Version 1, updated 30 July 2025, in Regional Plastics Outlook for Southeast and East Asia (DOI 10.1787/5a8ff43c-en). Values are native million tonnes (Mt), for the entire ASEAN Plus Three modeled region, never allocated to member countries. The 2022 modeled baseline and 2050 Baseline scenario endpoint are distinct from the three counterfactual 2022–2050 growth contributions. Contributions carry endpoint year 2050 for indexing; they are changes over the whole interval, not annual production or waste. Preserve negative contributions; the five rows must not be summed. No interpolation or scale conversion. Other figure workbooks and country/model outputs remain outside this limited extract.

Units: Metric tonnes for mass; percentages and tonnes/person where specified. Original values and units retained.

Preparation: Native values and dimensions preserved. Missing observations are not replaced with zero.

Open the original source · Retained records and definitions

OECD publication-specific terms; most postJuly2024 written contentCCBY4.0; verify dataset third-party rights. · Acquired 2026-09-30

OECD (2025), Regional Plastics Outlook for Southeast and East Asia, 2025edition. Retrieved 19 September 2026.

Acquisition note: All 27 map values for 2022 are retained. Absolute stocks, material/end-use breakdowns and historical series from the technical report remain outside this extract.

27 source observations. All 27 country values from the EEA map accompanying its 2026 material-stocks briefing. Reference year 2022; MISO model estimates of accumulated stocks, in tonnes per capita. Original Tonnes_per numbers and country/source identifiers retained without numerical conversion. Country polygons and their area/length metadata are omitted. Values are stocks per person, not annual extraction, waste, available secondary resources or recoverable tonnes. This edition remains separate from the older global MAT_STOCKS release. The map does not expose absolute stock values or a material/end-use breakdown; those remain outside this extract.

Units: Metric tonnes for mass; percentages and tonnes/person where specified. Original values and units retained.

Preparation: Native values and dimensions preserved. Missing observations are not replaced with zero.

Open the original source · Retained records and definitions

EEAreuse policy; underlyingBOKUMAT_STOCKSdataCCBY4.0, figure-specificterms apply. · Acquired 2026-09-30

European Environment Agency (2026), Material stocks in a circular economy, 2026report. Retrieved 19 September 2026.

2,824,314 source observations. 2024 HS22 chapters 01–27, 41, 44, 47–48, 50–53, 72–76 and 78–81. Scanned 11,250,411 original rows. Exporter partitions contain each reconciled bilateral observation once. Quantity is metric tonnes; missing quantity stays null; value is separately thousand USD. BACI reconciliation does not establish individual shipments. Excluded chapters are outside this retained material scope.

Units: Trade quantity q is metric tonnes. Trade value v is thousands of current USD; never used here to infer mass.

Preparation: Native values and dimensions preserved. Missing observations are not replaced with zero.

Open the original source · Retained records and definitions

Etalab Open Licence 2.0; attribution to BACI and CEPII required. · Acquired 2026-09-30

Gaulier, G. and Zignago, S. (2010). BACI: International Trade Database at the Product-Level. CEPII Working Paper 2010-23. Database release 202601.

Official Table 4 export records have been acquired for offline evaluation. Public quantitative reuse is withheld pending permission. The reporting manual instructs Parties to summarize exported/imported amounts by waste and D/R code; for general notifications it uses quantities received on movement documents, assigned by arrival date at the receiving facility. D/R codes indicate the operation the waste was destined for, not audited completion. Controlled waste includes hazardous and other waste.

Units: Reported Amount in metric tons, confirmed by Basel national-reporting manual Table 4. Numeric values retained in metric tonnes; original strings retained.

Preparation: Offline adapter preserves all source fields. Positive unambiguous mass strings are retained as tonnes. One declared-operation category per record; multiple codes remain unallocated. No facility join, mirror sum, or Comtrade crosswalk. Public artifacts are withheld.

Open the original source

Basel website Terms of Use, Introduction §1: permission limited to personal, non-commercial copying; no right to commercially redistribute, compile, or create derivative works. No separate quantitative dataset redistribution permission verified. https://www.basel.int/Home/TermsofUse/tabid/10656/Default.aspx · Acquired 2026-09-19

Basel Convention Secretariat. National reporting database, 2024 reports and control-tool documentation, retrieved 19 September 2026. Direction and metric-ton unit verified using the official 2025 national-reporting manual, Tables 4 and 5: https://www.basel.int/Portals/4/download.aspx?d=UNEP-CHW-NationalReporting-Manual-2025.English.docx

Acquisition note: All available annual 2025 HS2 quantity cells for EU27 reporters are retained. Other years, monthly data, detailed HS4/HS6/CN8 products and monetary/supplementary-unit measures remain outside this extract.

501,443 source observations. All available 2025 annual import/export quantity observations for 27 EU reporters, all 98 two-digit HS chapter codes in the pinned product list and all publisher-returned partners. Native QUANTITY_IN_100KG values retained in 100 kg units, without conversion. Chapter-level goods trade measures physical product quantities, not raw-material equivalents or embodied extraction. WORLD, intra/extra-EU/euro-area groups, confidential/special partner codes and individual partner countries are separate overlapping perspectives; never sum partner aggregates with countries. Imports and exports also overlap reports by other EU reporters and must not be added into a unique shipment ledger. Absent cells remain absent; explicit zeroes and publisher flags remain unchanged. HS2 chapters avoid overlap with HS4/HS6/CN8 rows; detailed products, other years, monthly series and monetary/supplementary-unit measures are outside this extract. Source JSON-stat cell indices, file hashes and publisher revision timestamps retained. No physical facilities, buyers or transport routes are inferred.

Units: Native units; no conversion applied because no numerical observations are bundled.

Preparation: Native values and dimensions preserved. Missing observations are not replaced with zero.

Open the original source · Retained records and definitions

Eurostat reuse with attribution, including commercial reuse, subject to its geographic and third-party exceptions: https://ec.europa.eu/eurostat/web/main/help/copyright-notice . Public outputs exclude non-EU/EFTA/official-candidate country records covered by those exceptions. · Acquired 2026-09-30

Eurostat. International trade in goods database (Comext), database and bulk-download documentation, retrieved 19 September 2026.

Candidate concordance and cross-check for plastic feedstocks, primary plastics, goods and scrap. Quantity coverage must be inspected independently of monetary coverage. Data-centre endpoint was not accessible in this environment; no observations acquired or shown.

Units: Native units; no conversion applied because no numerical observations are bundled.

Preparation: Catalogue and access/methodology review only. No data inserted into the atlas.

Open the original source

UNCTAD data terms; dataset-specific reuse conditions not verified in this acquisition. · Acquired 2026-09-19

UNCTAD. Global trade in plastics, UNCTADstat data centre.

Secondary access path and concordance resource, not an independent source to add to Comtrade totals. Bulk download and advanced queries may need an account; landing page returned HTTP403 during direct acquisition. No duplicate trade observations ingested.

Units: Native units; no conversion applied because no numerical observations are bundled.

Preparation: Catalogue and access/methodology review only. No data inserted into the atlas.

Open the original source

WITS and underlying provider conditions. Not a separate open licence for Comtrade data. · Acquired 2026-09-19

World Bank. World Integrated Trade Solution, UN Comtrade bulk-download documentation.

Alternative access path for the same Comtrade observations. Do not combine mirror totals as independent evidence. Explorer accessible; no separate numerical acquisition was needed because the current official Comtrade preview API supplied the selected physical flows.

Units: Native units; no conversion applied because no numerical observations are bundled.

Preparation: Catalogue and access/methodology review only. No data inserted into the atlas.

Open the original source

United Nations and underlying Comtrade terms. · Acquired 2026-09-19

UN Statistics Division. UNdata, ComTrade explorer.

Authoritative ISO-alpha2 / ISO-alpha3 / M49 lookup used to normalize Basel reporting country identifiers. Trade reporter codes must still follow Comtrade metadata: current India is 699 and United States 842, which differ from M49.

Units: Country identifiers, not mass or population values.

Preparation: Extract ISO2/ISO3 correspondence from the official HTML table, discarding headers and duplicate language rows. No material quantities derived.

Open the original source

United Nations website terms; country-code reference data with source attribution. · Acquired 2026-09-19

United Nations Statistics Division. Standard country or area codes for statistical use (M49), current table retrieved 19 September 2026.

Interpret legacy D/R codes only. D13–D15, R12 and R13 precede other operations. D8 and D9 yield material requiring further Section A disposal. Neither an R code nor a D code verifies treatment completion.

Units: Definitions and methodology, not quantitative observations.

Preparation: Short labels are editorial paraphrases. Exact native codes are retained; optional broad display groups are declared interpretations. More than one code remains an unallocated combination.

Open the original source

Official reference material; Basel website Terms of Use. Local research snapshots retained. No open data licence asserted. · Acquired 2026-09-19

Basel Convention (2025 revised text), Annex IV. Amendment decision BC-17/15 becomes effective on 1 January 2030.

Amounts are metric tons. Country of destination is the reported receiving country. For general notifications, sum facility-received quantities by arrival date. D/R columns describe the operation to which waste was destined.

Units: Definitions and methodology, not quantitative observations.

Preparation: No numeric transformation. Used to define normalized units and prevent the interface from claiming verified treatment.

Open the original source

Official reference material; Basel website Terms of Use. Local research snapshots retained. No open data licence asserted. · Acquired 2026-09-19

Secretariat of the Basel Convention (2025), Manual for completing the format for national reporting, Tables 4–5.

The amendment updates operation descriptions and adds R14, but becomes effective on 1 January 2030. It does not classify historical observations in this release.

Units: Definitions and methodology, not quantitative observations.

Preparation: Recorded for versioned taxonomy selection; no historical records changed.

Open the original source

Official reference material; Basel website Terms of Use. Local research snapshots retained. No open data licence asserted. · Acquired 2026-09-19

Basel Convention Secretariat, Amendment to Annex IV, decision BC-17/15.

Public bulk JSON archive. Exact series IDs, original units, source update dates and missing markers are retained. Display conversions only scale thousands of barrels/day to barrels/day and thousands of metric tonnes to tonnes.

Preparation: Public bulk JSON archive. Exact series IDs, original units, source update dates and missing markers are retained. Display conversions only scale thousands of barrels/day to barrels/day and thousands of metric tonnes to tonnes.

• Annual records may be estimated or revised. EIA bulk values do not provide observation-level estimate flags. Gas trade totals do not identify partners or separate LNG from pipeline deliveries.

Open the original source · Retained records and definitions

U.S. public domain; source acknowledgment required. Retained series declare copyright: None. · Acquired 2026-09-22

Primary oil annual CSV files, KBD only, for 2025 and 2026. KBD multiplied by 1,000 gives barrels/day. Original assessment codes and non-numeric markers are retained.

Preparation: Primary oil annual CSV files, KBD only, for 2025 and 2026. KBD multiplied by 1,000 gives barrels/day. Original assessment codes and non-numeric markers are retained.

• Monthly reporting is incomplete and revisions are possible. Missing values are not zero. Rates cannot be summed into monthly or annual volumes without weighting by days. Country trade totals do not identify shipping routes.

Open the original source · Retained records and definitions

Freely downloadable JODI data; cite IEF, JODI Oil World Database and access date. · Acquired 2026-09-22

All 9,074 source locations. Transform EPSG:3347 to WGS84; preserve provider, original ID, classification and status. Missing names are labeled by source class.

• Location records do not include waste throughput or verified downstream destinations.
• Several providers may describe the same physical site; counts are source records, not a deduplicated national facility census.
• The mixed category includes waste infrastructure whose exact role is unspecified.

Open the original source · Retained records and definitions

Open Government Licence – Canada · Acquired 2026-09-22

Transform EPSG:3347 to WGS84. Preserve native facility class, operating status, ownership and original ID. A representative point is used only for non-point facility geometry and labeled.

• Location and operating status describe the source snapshot, not current production.
• The source does not supply oil or gas throughput per facility.
• Provider overlap is retained; records are not claimed to be unique physical sites.

Open the original source · Retained records and definitions

Open Government Licence – Canada · Acquired 2026-09-22

Retain all source rows and all quality codes. Reported treatment, design capacity, unspecified reports and modeled effluent are separate measurement bases. Source population and treatment quality codes remain explicit.

• The dataset does not provide an observation year for each plant.
• Liquid volume cannot be added to solid-waste mass.
• Source estimates have no per-plant statistical interval.
• Outfall coordinates are estimated and distinct from plant coordinates.

Open the original source · Retained records and definitions

CC BY 4.0 · Acquired 2026-09-19

Ehalt Macedo et al. (2022), Distribution and characteristics of wastewater treatment plants within the global river network. https://doi.org/10.5194/essd-14-559-2022

Use the original area in km² and a representative point inside each mining polygon. Retain the source polygon ID. Derived mining records are shared under CC BY-SA 4.0.

• Footprints include pits, tailings, waste rock, ponds and processing areas.
• A footprint does not identify a mine operator, commodity, production or current operating status.

Open the original source · Retained records and definitions

CC BY-SA 4.0 · Acquired 2026-09-19

Maus et al. (2022), Global-scale mining polygons (Version 2). PANGAEA. https://doi.org/10.1594/PANGAEA.942325

Annual source short tons × exactly 0.90718474 = metric tonnes. Preserve each measurement year; missing intake or reporting year remains null.

• This is a voluntary program database, not a full US landfill census.
• Annual intake reports have different years and cannot be summed into a same-year national total.

Open the original source · Retained records and definitions

US government public data · Acquired 2026-09-19

Retain original tonnes/year and outfall coordinates. Countries absent in the source remain unassigned; no nearest-country substitution.

• These are model estimates, not observed shipments or measured discharges.
• Per-outfall uncertainty intervals and most river names are absent.
• Most country assignments are unavailable in the source.

Open the original source · Retained records and definitions

CC BY 4.0 · Acquired 2026-09-19

Meijer et al. (2021), More than 1000 rivers account for 80% of global riverine plastic emissions into the ocean. https://doi.org/10.1126/sciadv.aaz5803

Retain every source row, native plant ID, primary and secondary fuels, capacity in MW, capacity observation year, original provider URLs and all annual generation fields. Reported GWh and modeled GWh remain separate original fields.

• Capacity in MW is a rated power level, not annual energy production.
• Current operating status and recent additions or closures are not supplied.
• Location and capacity accuracy vary by contributing provider.
• Primary fuel does not establish a complete annual fuel mix. Generation may use calendar, fiscal or regulatory years.

Open the original source · Retained records and definitions

CC BY 4.0 · Acquired 2026-09-22

World Resources Institute, Global Power Plant Database. Archived public repository snapshot retrieved 22 September 2026. CC BY 4.0.

Retain native point-source centroids, identifiers and 2025 inventory estimates. CO₂e uses IPCC AR6 100-year global warming potentials. Upstream activity fields marked license restricted are unavailable, not zero. Original API pages and hashes are retained. The release 15 extension removes only byte-equivalent repeated source IDs from API pagination. All original pages and hashes remain available.

• Emissions combine models and public reports. The summary API does not provide a per-record measurement method or uncertainty interval.
• Tonnes of CO₂e describe climate impact, not tonnes of waste handled or material produced.
• Point-source centroids can represent a production area or complex; they are not guaranteed building-level coordinates.
• Names may be descriptive labels assigned by the provider. Overlap with other source inventories is retained.
• Selected subsectors and source coverage do not form a complete census of operating facilities.
• Offset pagination repeats some identical rows at tied sort values. Identical IDs are deduplicated; conflicting versions stop the import. Completeness beyond the retained API response is not claimed.

Open the original source · Retained records and definitions

CC BY 4.0; listed external-source exceptions · Acquired 2026-09-22

Climate TRACE. Emissions inventory, 2025 annual CO₂e estimates. API v7 snapshot retrieved 22 September 2026. Modified by filtering point-source records, organizing subsectors and labeling measurement basis; quantities and coordinates unchanged.

Production series, original units, flags, notes and source row references. Countries and native world totals retained separately.

Units: Tonnes or cubic metres, depending on the product

• Inputs and outputs overlap. Do not add production stages.
• Cotton lint ends in 2023 in this edition.
• Source estimates and imputation flags are retained.

Open the original source

CC BY 4.0; FAO terms apply · Acquired 2026-09-22

Monthly sums of the public daily port series, grouped by source port ID using ArcGIS statistical queries. Dates, day counts and original port identifiers are retained. Ship classes are container, dry bulk, general cargo, roll-on/roll-off and tanker. Calls are visits, not unique ships.

Units: Port calls (visits); source-modeled metric tonnes of cargo; WGS84 coordinates

Preparation: Monthly sums of the public daily port series via source-side statistical queries. Preserve observed-day counts, original IDs, dates, query URLs and raw response hashes.

• Port calls are derived from AIS and port boundaries. Source transit filters and AIS coverage affect the counts.
• Cargo tonnes are modeled from vessel draft, deadweight and payload changes; they are not weighed cargo or customs declarations.
• The vessel class does not establish the commodity, origin, destination or owner of a shipment.
• A port location may represent a port complex or offshore terminal. Records are not individual berths.
• This is a reviewed historical snapshot, not a live AIS feed.

Open the original source · Retained records and definitions

IMF statistical-data terms; attribution and transformation disclosure required · Acquired 2026-09-23

Sources: UN Global Platform; IMF PortWatch (portwatch.imf.org).

Source hourly AIS position reports, including stationary vessels. Source overviews averaged to a display grid, reprojected to Web Mercator, then averaged for lower zoom levels. Color is a relative visual index, not a quantity of ships or cargo.

Units: Relative visual intensity; native source counts hourly position reports per 0.005-degree cell

Preparation: Log-scaled relative density; clipped at the positive-cell 99.5th percentile; display raster is 4,096 pixels wide.

• Historical activity, not live vessel positions.
• Generalized to a 4,096-pixel-wide world raster; do not interpret individual berths or vessels.
• AIS coverage varies. Blank cells do not establish that no shipping occurred.
• Activity is not specific to the material selected in OVERSHOOT.

Open the original source · Retained records and definitions

CC BY 4.0 · Acquired 2026-09-23

World Bank, Global Shipping Traffic Density, commercial shipping layer. Based on IMF hourly AIS analysis, 2015–2021. Display generalized by OVERSHOOT.

218148 observations. Complete four satellite regional account tables (consumption, production, imports and exports) from the public industry-by-industry 2019 archive, version 3.8.1. Native units include kt, kg, Mm3, km2, TJ and monetary/labour units; no scale conversion or addition across stressors. Used and unused extraction, ore mass and emissions remain distinct. 2019 is the modeled account year, not the latest observed input year: materials are nowcast after 2013; land/water and most others after 2011. Rest-of-world regions retain EXIO identities and are not allocated to nations. Accounts represent modeled embodied burdens, not observed shipments. The 3.10.2 catalogue source has different terms and is not imported through this historical release. Changes: native TSV cells converted to country-partitioned JSON with original member/row/column lineage; no numerical transformation. Full source archive and transformation code remain available for attribution/share-alike compliance.

Open the original source · Retained records and definitions

CC BY-SA 4.0; this derived extract retains CC BY-SA 4.0. Attribution: Stadler et al., EXIOBASE 3, version 3.8.1, DOI 10.5281/zenodo.4588235. · Acquired 2026-09-30

315,758 observations. Complete acquired monthly CSV. Natural gas million m³, terajoules and LNG thousand tonnes are distinct units. Total imports/exports overlap their LNG and pipeline components. Stock changes may be negative. Missing source markers remain null and retain their original text. Assessment codes are not statistical confidence intervals.

Open the original source · Retained records and definitions

Freely downloadable; cite IEF/JODI Gas World Database and access date · Acquired 2026-09-30

26,563 observations. Seven selected harmonized measures, 2000 onward. Native TWh are energy content, not material mass. Country aggregates excluded. Underlying variable sources are documented in the retained codebook.

Open the original source · Retained records and definitions

OWID CC BY 4.0; underlying EIA/Energy Institute and other provider terms apply · Acquired 2026-09-30

833 observations. Native JSON-stat dimensions, units and status flags. Regional aggregates are separately prefixed ESTAT and excluded from country maps. No unit conversion or addition to IRP accounts. Kosovo cells withheld under publisher exception.

Open the original source · Retained records and definitions

Eurostat reuse policy; geographic and third-party exceptions · Acquired 2026-09-30